Initial Search -> 454 articles identified
Titles and Abstracts -> 24 articles relevant
Detailed Review -> 3 articles
Exclude one more article that reported association between glucose levels and
outcome but with no difference in treatments.
Final Evidence -> 2 articles
Article: Resuscitation Feb--2008; 76 (2): 214-20. Strict normoglycaemic blood glucose levels
in the therapeutic management of patients within 12h after cardiac arrest might not be
necessary.. Losert, Heidrun; Sterz, Fritz; Roine, Risto O; Holzer, Michael; Martens, Patrick;
Cerchiari, Erga; Tiainen, Marjaana; Müllner, Marcus; Laggner, Anton N; Herkner, Harald;
Bischof, Martin G;
The admission blood glucose level after cardiac arrest is predictive of outcome. However the blood
glucose levels in the post-resuscitation period, that are optimal remains a matter of debate. We
wanted to assess an association between blood glucose levels at 12h after restoration of spontaneous
circulation and neurological recovery over 6 months.A total of 234 patients from a multi-centre trial
examining the effect of mild hypothermia on neurological outcome were included. According to the
serum glucose level at 12h after restoration of spontaneous circulation, quartiles (Q) were generated:
Median (range) glucose concentrations were for QI 100 (67-115 mg/dl), QII 130 (116-143 mg/dl),
QIII 162 (144-193 mg/dl) and QIV 265 (194-464 mg/dl).In univariate analysis there was a strong
non-linear association between blood glucose and good neurological outcome (odds ratio compared
to QIV): QI 8.05 (3.03-21.4), QII 13.41 (4.9-36.67), QIII 1.88 (0.67-5.26). After adjustment for sex, age,
"no-flow" and "low-flow" time, adrenaline (epinephrine) dose, history of coronary artery disease and
myocardial infarction, and therapeutic hypothermia, this association still remained strong: QI 4.55
(1.28-16.12), QII 13.02 (3.29-49.9), QIII 1.37 (0.38-5.64).There is a strong non-linear association of
survival with good neurological outcome and blood glucose levels 12h after cardiac arrest even after
adjusting for potential confounders. Not only strict normoglycaemia, but also blood glucose levels
from 116 to 143 mg/dl were correlated with survival and good neurological outcome, which might
have an important therapeutic implication.
DOI: http://dx.doi.org/10.1016/j.resuscitation.2007.08.003
http://www.ncbi.nlm.nih.gov/pubmed?term=17870226
Article: Intensive Care Med Dec--2007; 33 (12): 2093-100. Strict versus moderate glucose
control after resuscitation from ventricular fibrillation.. Oksanen, Tuomas; Skrifvars, Markus
B; Varpula, Tero; Kuitunen, Anne; Pettilä, Ville; Nurmi, Jouni; Castrén, Maaret;
Elevated blood glucose is associated with poor outcome in patients resuscitated from out-of-hospital
cardiac arrest (OHCA). Our aim was to determine whether strict glucose control with intensive
insulin treatment improves outcome of OHCA patients.A randomized, controlled trial.Two university
hospital intensive care units.Ninety patients resuscitated from OHCA with ventricular fibrillation
detected as the initial rhythm were treated with therapeutic hypothermia.Patients were randomized
into two treatment groups: a strict glucose control group (SGC group), with a blood glucose target of
4-6 mmol/l, or a moderate glucose control group (MGC group), with a blood glucose target of 6-8
mmol/l. Both groups were treated with insulin infusion for 48 h, because a control group with no
treatment was considered unethical.Baseline data were similar in both groups. In the SGC group 71%
of the glucose measurements were within the target range compared with 41% in the MGC group.
Median glucose was 5.0 mmol/l in the SGC group and 6.4 mmol/l in the MGC group. The occurrence
of moderate hypoglycemic episodes was 18% in the SGC group and 2% in the MGC group (p = 0.008).
No episodes of severe hypoglycemia occurred. Mortality by day 30 was 33% in the SGC group and
35% in the MGC group (p = 0.846); the difference was 2% (95% CI -18% to +22%).We found no
additional survival benefit from strict glucose control compared with moderate glucose control with
a target between 6 and 8 mmol/l in OHCA patients.
http://www.ncbi.nlm.nih.gov/pubmed?term=17928994
Criteria
Judgeme
nts
○ No
○
Probably
no
Problem
Is there a
problem
priority?
○
Uncertain
●
Additional
considerati
ons
Research evidence
One RCT of 90 subjects found no reduction in 30-day
mortality (RR 0.94; 95%CI 0.53-1.68) when subjects
were assigned to strict (4-6mmol/L ) versus moderate
(6-8 mmol/L) glucose control. One before-and-after
observational study of 119 subjects found reduced inhospital mortality (RR 0.46; 95%CI 0.28-0.76) after
implementation of a bundle of care that included
defined glucose management (5-8 mmil/L), but the
isolated effect of glucose management cannot be
separated from the effect of other parts of the bundle.
Probably
yes
○ Yes
○ Varies
○ No
included
studies
What is
the
overall
certainty
of this
evidence?
○ Very
The relative importance or values of the main
outcomes of interest:
Outcome
low
● Low
○
○ High
Outco
me
○
Is there
important
uncertaint
y about
how much
people
value the
main
outcomes
?
Important
uncertaint
y or
variability
○
Possibly
important
uncertaint
y or
variability
○
Probably
no
important
uncertaint
y of
variability
● No
Certainty of the
evidence (GRADE)
Summary of findings: other target range
Moderate
Benefits
& harms
of the
options
Relative
importance
Without a
specific
target
range for
blood
glucose
managem
ent (eg.
strict 4-6
mmol/L)
With a
specific
target
range for
blood
glucose
managem
ent (eg.
strict 4-6
mmol/L)
Differe
nce
(95%
CI)
Relati
ve
effect
(RR)
(95%
CI)
Criteria
Judgeme
nts
important
uncertaint
y of
variability
○ No
known
undesirabl
e
○ No
●
Probably
no
Are the
desirable
anticipate
d effects
large?
○
Uncertain
○
Probably
yes
○ Yes
○ Varies
○ No
○
Are the
undesirabl
e
anticipate
d effects
small?
Probably
no
○
Uncertain
●
Probably
yes
○ Yes
○ Varies
Are the
desirable
effects
large
relative to
undesirabl
e effects?
○ No
●
Probably
no
○
Uncertain
○
Probably
Research evidence
Additional
considerati
ons
Criteria
Judgeme
nts
Research evidence
yes
○ Yes
○ Varies
○ No
●
Strict glycemic control is labor intensive and requires
increased monitoring.
Probably
no
Are the
resources
required
small?
○
Uncertain
○
Probably
yes
○ Yes
○ Varies
Resource
use
○ No
●
Is the
increment
al cost
small
relative to
the net
benefits?
Probably
no
○
Uncertain
○
Probably
yes
○ Yes
○ Varies
○
Increased
○
Equity
What
Probably
would be
the impact increased
on health
inequities
Uncertain
?
●
○
Probably
reduced
○
Labor costs may be significant
Additional
considerati
ons
Criteria
Judgeme
nts
Research evidence
Additional
considerati
ons
Reduced
○ Varies
○ No
○
Is the
option
Acceptabi acceptabl
lity
e to key
stakehold
ers?
Probably
no
○
Uncertain
●
Probably
yes
○ Yes
○ Varies
○ No
○
Probably
no
○
Feasibilit
y
Is the
option
Uncertain
feasible to
implement
?
Probably
yes
●
○ Yes
○ Varies
Recommendation
Should a specific target range for blood glucose management (eg. strict
4-6 mmol/L) vs. other target range be used for adults with ROSC after
cardiac arrest in any setting?
Balance of
consequences
Undesirable
consequences
clearly
outweighdesirable
consequences in
most settings
Undesirable
consequences
probably
outweigh
desirable
consequences
in most
The balance
between
desirable and
undesirable
consequences
is closely
balanced or
Desirable
consequences
probably
outweigh
undesirable
consequences
in most
Desirable
consequences
clearly
outweigh
undesirable
consequences
in most
settings
uncertain
settings
settings
○
●
○
○
○
Type of
recommendation
We recommend
against offering this
option
We suggest not
offering this
option
We suggest
offering this
option
We recommend
offering this option
○
●
○
○
Recommendation
We suggest not selecting any specific target range of glucose management versus
any other target range in adults with ROSC after cardiac arrest.
Justification
One RCT of 90 subjects found no reduction in 30-day mortality (RR 0.94; 95%CI
0.53-1.68) when subjects were assigned to strict (4-6mmol/L ) versus moderate
(6-8 mmol/L) glucose control. One before-and-after observational study of 119
subjects found reduced in-hospital mortality (RR 0.46; 95%CI 0.28-0.76) after
implementation of a bundle of care that included defined glucose management (5-8
mmil/L), but the isolated effect of glucose management cannot be separated from
the effect of other parts of the bundle.
Subgroup
considerations
Many observational studies not included in the final analysis noted that elevated
glucose levels are more common in patients with more severe injuries even prior to
treatment.
Implementation
considerations
Strict glycemic control is labor intensive. In other populations, implementation of
strict glycemic control is associated with increased episode of hypoglycemia, which
might be detrimental.
Monitoring and
evaluation
Hypoglycemic episodes should be monitored.
Research
possibilities
There is no trial comparing glucose management to no glucose
management. There is incomplete data about different ranges of glucose
control. There are no data about glucose ranges >8 mmol/L.